强跟踪有限微分扩展卡尔曼滤波算法在异步电机中的应用

The Application of Strong Tracking Finite-Difference Extended Kalman Filter Algorithm in Induction Motor

针对异步电机直接转矩控制低速转矩脉动大的问题,提出了一种自适应强跟踪有限微分扩展卡尔曼滤波算法(STFDEKF)。该算法采用多项式近似技术和一阶中心差分法计算非线性函数的偏导数,它具有二阶非线性近似的能力;同时引入强跟踪因子来修改状态的先验协方差矩阵。文中将STFDEKF算法扩展至感应电机参数辨识,设计了针对非线性系统的STFDEKF滤波器;同时对五个电机变量进行观测,并将观测到的转速信息引入至感应电机控制系统,组成无速度传感器控制闭环。仿真结果表明,STFDEKF算法相比扩展卡尔曼算法(EKF)和有限微分扩展卡尔曼算法(FDEKF)虽然计算过程复杂,但是该算法具有更好地跟踪能力和滤波可靠性,所以在对磁链和转速目标进行跟踪时,STFDEKF算法是一种很有效的算法。

To solve the problem of tracking the stator flux amplitude and angle of the induction motor（IM）,an adaptive algorithm,strong tracking finite-difference extended Kalman filter（STFDEKF）,was proposed for signal tracking of the IM,using direct torque control method.This method used polynomial approximations obtained with a sterling interpolation formula to approximate the derivative of the nonlinear function,and used strong tracking factors to modify the prior covariance matrix.The proposed algorithm improved the tracking accuracy,enlarges the applied area and enhanced the filtering convergence.The performance of the proposed algorithm was compared with that of the extended Kalman filter（EKF） and the finite-difference extended Kalman filter（FDEKF） using a Monte Carlo simulation.The simulation results show that STFDEKF outperformed EKF and FDEKF in terms of tracking accuracy and filter credibility,although it had higher computational cost.It was concluded that the STFDEKF is an effective algorithm for the flux and speed target tracking problem.